CN116566914A - Bypass TCP acceleration method, device, equipment and medium - Google Patents

Abstract

The present invention relates to the field of Internet technology, and provides a bypass TCP acceleration method, device, equipment and medium, which are applied to acceleration equipment or acceleration services deployed by bypass, and improve the link state without affecting the original TCP link. And the deployment is flexible; capture data packets in real time, and distribute the captured data packets to different threads to improve concurrent processing efficiency; when packet loss is detected, the data packet is retransmitted in combination with the cache, without the need to resend the packet from the sender , reducing the time required for resending packets, and replenishing packets in advance before packet loss affects the link to ensure smooth links; when congestion is detected in the data stream corresponding to the data packet, congestion control is performed to reduce the occurrence of link congestion Possibility, improve the bandwidth utilization of the link; detect whether the data packet needs to be confirmed in advance, and reduce the overall delay and packet loss rate of the link by replying ACK in advance and resending the data packet.

Description

Bypass TCP acceleration method, device, equipment and medium

technical field

The invention relates to the technical field of the Internet, in particular to a bypass TCP acceleration method, device, equipment and medium.

Background technique

TCP (Transmission Control Protocol, Transmission Control Protocol) is a connection-oriented, reliable, byte-stream-based transport layer communication protocol, which relies on sequential numbering of data packets and confirmation mechanism, in an unreliable network environment Provide reliable transmission guarantee for the upper layer protocol.

Although TCP is a reliable protocol, problems such as instability and large fluctuations may occur under some special network conditions, such as long-distance network links or wireless networks. Taking wireless networks as an example, wireless networks use radio waves as a transmission medium to extend data transmission and communication functions between devices without physical connections, allowing devices to interconnect and communicate wirelessly without using traditional wired connections , such as cables or optical fibers. However, the wireless network also brings problems of high bit error rate and unstable network conditions. When the user’s network environment is poor, a large number of lost packets will be generated. These lost packets need to be retransmitted from the sender, which has a great impact full link performance.

In view of the instability and large fluctuations of TCP under special network conditions, the following solutions are usually adopted in the prior art:

1. Increase the number of intermediate routes and switches to improve signal coverage.

However, this method is expensive and consumes a lot.

2. Deploy proxy services on intermediate routers and switches to split TCP connections.

However, this method generally can only accelerate the slow start of TCP, and will occupy a large amount of device resources and consume a lot of resources. In addition, intermediate routes need to be modified, which will affect existing services and pose a high risk. At the same time, this solution splits the original TCP connection into two, destroying the original TCP link. Once there is a problem with the proxy service, the entire TCP link will be paralyzed, and the configuration is cumbersome.

Contents of the invention

In view of the above, it is necessary to provide a bypass TCP acceleration method, device, equipment and medium, aiming to solve the problem of slow packet loss and retransmission and high delay in an unstable and fluctuating network environment, as well as the traditional proxy pair The problem that the original TCP link has a large impact, and solves the problem that the traditional proxy configuration is cumbersome, occupies a large amount of resources, and deployment or modification will affect the original business.

A bypass TCP acceleration method, applied to bypass deployed acceleration equipment or acceleration services, the bypass TCP acceleration method includes:

Capture data packets in real time, and divert the captured data packets to corresponding threads when capturing data packets;

Detecting whether there is a risk of data packet loss, and when detecting that there is a risk of data packet loss, retransmitting the data packet in combination with the buffer, and retransmitting the data packet to the receiver of the data packet;

During the retransmission process, detect whether the data flow corresponding to the data packet is congested, and perform congestion control when it is detected that the data flow corresponding to the data packet is congested;

After performing congestion control, detecting whether the data packet needs to be confirmed in advance;

When it is detected that the data packet needs to be acknowledged in advance, generate a simulated ACK packet of the data packet, and transmit the simulated ACK packet to the sender of the data packet.

According to a preferred embodiment of the present invention, the method further includes:

Obtain the intermediate device of the sender and the receiver, and obtain the free port of the intermediate device, configure mirroring for the free port, and establish a connection with the mirror server through the free port, and in the mirror server Deploy the acceleration device on; or

The intermediate devices of the sender and the receiver are obtained, and the acceleration service is deployed on the intermediate devices.

According to a preferred embodiment of the present invention, said shunting captured data packets to corresponding threads includes:

Get the header format of each data packet;

Determine the protocol type of each data packet according to the header format of each data packet;

Analyze the TCP/IP quadruple of each data packet based on the protocol type of each data packet, and obtain the analysis result;

Divide each data packet into a corresponding TCP flow according to the analysis result;

Configure a corresponding thread for each data packet according to the TCP flow to which each data packet belongs.

According to a preferred embodiment of the present invention, before performing the retransmission of the data packet in combination with the cache, the method further includes:

Obtain the MTU determined when the sender and the receiver perform a TCP handshake;

Segmenting unacknowledged data packets with the length of the MTU, and storing the segmented data packets in the cache in the form of an ordered two-dimensional linked list;

Wherein, when receiving the ACK packet of any data packet in the cache, deleting the field of the interval to which the arbitrary data packet belongs from the cache;

Wherein, when querying the lost data packet in the cache, a binary search is performed within a preset length range at the end of the ordered two-dimensional linked list.

According to a preferred embodiment of the present invention, the detecting whether there is a risk of data packet loss, and when it is detected that there is a risk of data packet loss, performing retransmission of the data packet in conjunction with caching includes:

Obtain the bandwidth of the TCP stream to which each data packet belongs, the historical packet loss rate, and the storage duration in the cache;

Acquiring a first weight corresponding to the bandwidth, a second weight corresponding to the historical packet loss rate, and a third weight corresponding to the storage duration;

Calculate the product of the bandwidth and the first weight to obtain a first value, calculate the product of the historical packet loss rate and the second weight to obtain a second value, and calculate the storage duration and the third weight The product gets the third value;

calculating the sum of the first value, the second value and the third value to obtain the current dynamic packet loss rate of each data packet;

Obtain the pre-configured packet loss rate threshold;

When the dynamic packet loss rate corresponding to a data packet is greater than the packet loss rate threshold, it is determined that the data packet has a loss risk;

Obtain the SACK field from the TCP header corresponding to the data packet;

Obtain the sequence number of the lost data packet according to the SACK field;

Querying and obtaining the missing fields from the cache according to the sequence number of the data packet;

Retransmit the lost field according to the data packet sequence number.

According to a preferred embodiment of the present invention, the method further includes:

In the process of retransmitting the lost field according to the sequence number of the data packet, configuring a timer for each retransmitted lost field;

Retransmitting the lost fields for each retransmission based on the timer with a timeout.

According to a preferred embodiment of the present invention, the detecting whether the data flow corresponding to the data packet is congested, and performing congestion control when detecting that the data flow corresponding to the data packet is congested includes:

The congestion index is calculated using the following formula:

;

Wherein, R represents the pre-configured theoretical bandwidth, S represents the size of the data packet, t _RTT represents the round-trip delay of the data packet, t _RTO represents the pre-configured transmission overtime limit, and p represents the historical packet loss rate, b represents the number of confirmation packets of an ACK; Y represents the congestion index, bw represents the bandwidth of the TCP flow to which the data packet belongs; cwnd represents the size of the congestion window;

Get the pre-configured congestion index threshold;

When the congestion index Y is less than the congestion index threshold, it is determined that the TCP flow corresponding to the data packet is congested;

reducing the sending frequency of the analog ACK packet used for early confirmation in the TCP flow, and increasing the retransmission frequency of the data packet corresponding to the TCP flow in the cache;

Wherein, for other TCP flows with the same IP, reduce the sending frequency of the simulated ACK packets used for early confirmation in the other TCP flows, and increase the retransmission frequency of the data packets corresponding to the other TCP flows in the cache.

According to a preferred embodiment of the present invention, the detecting whether the data packet needs to be confirmed in advance includes:

Obtain the pre-configured fourth weight, fifth weight and sixth weight;

Calculate the product of the bandwidth of the TCP stream to which the data packet belongs and the fourth weight to obtain a fourth value, calculate the product of the historical packet loss rate and the fifth weight to obtain a fifth value, and calculate the congestion index multiplied by said sixth weight to obtain a sixth value;

calculating the sum of the fourth value, the fifth value and the sixth value to obtain a target value;

Get pre-configured preset thresholds;

When the target value is greater than the preset threshold, it is determined that the data packet needs to be acknowledged in advance.

A bypass TCP acceleration device, running on the acceleration device or acceleration service deployed by the bypass, the bypass TCP acceleration device includes:

The distribution unit is used to capture the data packets in real time, and when the data packets are captured, the captured data packets are distributed to the corresponding threads;

A detection unit, configured to detect whether there is a risk of data packet loss, and when it is detected that there is a risk of data packet loss, retransmit the data packet in combination with the cache, and retransmit the data packet to the data the recipient of the packet;

A congestion control unit, configured to detect whether the data flow corresponding to the data packet is congested during the retransmission process, and perform congestion control when it is detected that the data flow corresponding to the data packet is congested;

The detection unit is further configured to detect whether the data packet needs to be confirmed in advance after performing congestion control;

The retransmission unit is configured to generate a simulated ACK packet of the data packet when it is detected that the data packet needs to be acknowledged in advance, and transmit the simulated ACK packet to the sender of the data packet.

A computer device comprising:

a memory storing at least one instruction; and

A processor, executing instructions stored in the memory to implement the bypass TCP acceleration method.

A computer-readable storage medium, at least one instruction is stored in the computer-readable storage medium, and the at least one instruction is executed by a processor in a computer device to implement the bypass TCP acceleration method.

It can be seen from the above technical solutions that the present invention is applied to the acceleration device or acceleration service deployed by the bypass, and improves the link status without affecting the original TCP link. The effect of splitting the TCP link into two segments (that is, from the sender to the deployment node of the acceleration device or acceleration service, and from the deployment node of the acceleration device or acceleration service to the receiver), limits the impact of packet loss to any one of the links In this way, the lost data packets can be retransmitted faster, and the deployment is flexible; the data packets are captured in real time, and the captured data packets are distributed to different threads to improve the concurrent processing efficiency; when the data packet loss is detected, Combined with the cache to retransmit the data packet, there is no need to resend the packet from the sender, just get the lost packet from the intermediate acceleration cache and retransmit it to the receiver, which reduces the time required for resending the packet, and advances before the packet loss affects the link Supplement the packet to ensure the smooth link; perform congestion control when the data stream corresponding to the data packet is detected to be congested, which reduces the possibility of link congestion and improves the bandwidth utilization of the link; detects whether the data packet needs to be forwarded Confirm that by replying ACK in advance and resending data packets, the equivalent two-stage rate from the sender to the deployment node of the acceleration device or acceleration service, and from the deployment node of the acceleration device or acceleration service to the receiver is balanced, reducing the link speed. Overall latency and packet loss.

Description of drawings

Fig. 1 is a flowchart of a preferred embodiment of the bypass TCP acceleration method of the present invention.

FIG. 2 is a schematic diagram of hardware deployment of the acceleration device of the present invention.

Fig. 3 is a schematic diagram of software deployment of the acceleration service of the present invention.

Fig. 4 is a functional block diagram of a preferred embodiment of the bypass TCP acceleration device of the present invention.

Fig. 5 is a schematic structural diagram of a computer device in a preferred embodiment of the bypass TCP acceleration method of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

As shown in FIG. 1 , it is a flow chart of a preferred embodiment of the bypass TCP acceleration method of the present invention. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.

The bypass TCP acceleration method is applied to one or more computer devices, and the computer device is a device that can automatically perform numerical calculation and/or information processing according to preset or stored instructions, and its hardware includes but It is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

The computer device may be any electronic product capable of man-machine interaction with the user, for example, a personal computer, a tablet computer, a smart phone, a personal digital assistant (Personal Digital Assistant, PDA), a game console, an interactive Internet TV ( Internet Protocol Television, IPTV), smart wearable devices, etc.

The computer equipment may also include network equipment and/or user equipment. Wherein, the network device includes, but is not limited to, a single network server, a server group composed of multiple network servers, or a cloud composed of a large number of hosts or network servers based on cloud computing (Cloud Computing).

The server can be an independent server, or it can provide cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, content distribution network (ContentDelivery Network) , CDN), and cloud servers for basic cloud computing services such as big data and artificial intelligence platforms.

Among them, artificial intelligence (AI) is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. .

Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, and mechatronics. Artificial intelligence software technology mainly includes computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning/deep learning.

The network where the computer device is located includes, but is not limited to, the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (Virtual Private Network, VPN) and the like.

This embodiment is applied to acceleration devices or acceleration services deployed in bypass, including:

S10, capture data packets in real time, and when the data packets are captured, distribute the captured data packets to corresponding threads.

In this embodiment, the method also includes:

(1) Obtain the intermediate device of the sender and the receiver, and obtain the free port of the intermediate device, configure mirroring for the free port, and establish a connection with the mirror server through the free port, and Deploying the acceleration device on the mirror server;

Wherein, the intermediate device may include, but is not limited to: an intermediate router, switch, gateway, base station, and the like.

Please refer to FIG. 2 , which is a schematic diagram of hardware deployment of the acceleration device of the present invention. Wherein, for the intermediate device between the sender and the receiver, the connection between the intermediate device and the mirror server is established, and the acceleration device is deployed on the mirror server. Under normal circumstances, the sender sends a data packet to the receiver through the intermediate device, and the receiver feeds back the real ACK packet to the sender through the intermediate device after receiving the data packet. The acceleration device acts as a proxy device deployed in the bypass to send simulated ACK packets to the sender and retransmitted data packets to the receiver.

In the foregoing embodiments, hardware deployment of the acceleration device is implemented.

(2) Obtain the intermediate devices of the sender and the receiver, and deploy the acceleration service on the intermediate devices.

Please refer to FIG. 3 , which is a schematic diagram of software deployment of the acceleration service of the present invention. Wherein, the acceleration service is deployed on the intermediate device. Under normal circumstances, the sender sends a data packet to the receiver through the intermediate device, and the receiver feeds back the real ACK packet to the sender through the intermediate device after receiving the data packet. The acceleration service is deployed as a proxy service in the bypass, and is used to send simulated ACK packets to the sender and retransmitted data packets to the receiver.

In the above embodiments, the software deployment of the acceleration service is realized.

Specifically, hardware deployment or software deployment can be arbitrarily selected according to actual needs, and the original network architecture system does not need to be modified during deployment, and maintenance costs will not be increased.

Through the above embodiments, the bypass deployment of acceleration devices or acceleration services can be realized, which solves the problem of cumbersome configuration of traditional agents, large resource occupation, deployment or modification that will affect the original business, and the impact on the original TCP (Transmission Control Protocol, transmission control protocol) chain. The road has a big impact on the problem.

In this embodiment, the splitting the captured data packets to corresponding threads includes:

Get the header format of each data packet;

Determine the protocol type of each data packet according to the header format of each data packet;

Analyze the TCP/IP (Transmission Control Protocol/Internet Protocol, Transmission Control Protocol/Internet Protocol) quadruple of each data packet based on the protocol type of each data packet, and obtain the analysis result;

Divide each data packet into a corresponding TCP flow according to the analysis result;

Configure a corresponding thread for each data packet according to the TCP flow to which each data packet belongs.

Wherein, the protocol type of each data packet may include Internet Protocol Version 4 (Internet Protocol Version 4, IPv4), Internet Protocol Version 6 (Internet Protocol Version 6, IPv6) and the like.

Through the foregoing embodiments, different data packets are divided into different threads for processing, which can effectively improve data processing efficiency.

Furthermore, dynamic load balancing can also be performed according to the load pressure change of each thread, so as to make full use of multi-core performance.

S11. Detect whether there is a risk of data packet loss, and when it is detected that there is a risk of data packet loss, retransmit the data packet in combination with buffering, and retransmit the data packet to the receiver of the data packet square.

In this embodiment, before performing the retransmission of the data packet in combination with the cache, the method further includes:

Obtain the MTU (Maximum Transmission Unit, maximum transmission unit) determined when the sender and the receiver perform a TCP handshake;

The unacknowledged data packets are divided according to the length of the MTU, and the divided data packets are stored in the cache in the form of an ordered two-dimensional linked list.

In the above embodiment, since the network card has mechanisms such as TSO (TCP Segmentation Offload, TCP segment offload), GSO (Generic Segmentation Offload, delayed fragmentation technology), the data packets will be merged, and the length of the captured data packets will often be greater than the MTU. In order to accurately resend the lost data segment, this embodiment divides the data packet into MTU lengths for storage, and modifies the serial number of the data packet header at the same time.

Both TSO and GSO technologies move the segmentation and merging operations of data packets from the CPU to the network card hardware, thereby reducing the load on the CPU.

MTU refers to the size of the largest data packet that can be transmitted in network communication.

Wherein, when an ACK (Acknowledge character, acknowledgment character) packet of any data packet in the cache is received, the field of the interval to which the arbitrary data packet belongs is deleted from the cache.

For example: detect the flag bit of the TCP data header, if it has an ACK field (used to confirm whether the corresponding data packet is confirmed), then transfer the ACK field to the cache, so as to detect whether the corresponding data packet has been received through the ACK field be confirmed. Furthermore, removing the confirmed data packets in the cache and clearing useless data packet caches in time can greatly reduce the memory usage. Furthermore, it is detected whether the data packet contains data, and if it contains data, the data packet is cached for retransmission that may occur later.

Wherein, when querying the lost data packet in the cache, a binary search is performed within a preset length range at the end of the ordered two-dimensional linked list.

For example: binary search can be performed within 20% of the length of the end of the ordered two-dimensional linked list to reduce the search range, and combined with binary search, the number of queries is reduced, CPU consumption is reduced, and query efficiency is improved.

In this embodiment, the detecting whether there is a risk of data packet loss, and when it is detected that there is a risk of data packet loss, retransmitting the data packet in conjunction with caching includes:

Obtain the bandwidth of the TCP stream to which each data packet belongs, the historical packet loss rate, and the storage duration in the cache;

Acquiring a first weight corresponding to the bandwidth, a second weight corresponding to the historical packet loss rate, and a third weight corresponding to the storage duration;

Calculate the product of the bandwidth and the first weight to obtain a first value, calculate the product of the historical packet loss rate and the second weight to obtain a second value, and calculate the storage duration and the third weight The product gets the third value;

calculating the sum of the first value, the second value and the third value to obtain the current dynamic packet loss rate of each data packet;

Obtain the pre-configured packet loss rate threshold;

When the dynamic packet loss rate corresponding to a data packet is greater than the packet loss rate threshold, it is determined that the data packet has a loss risk;

Obtain the SACK (Selective ACK) field from the TCP header corresponding to the data packet;

Obtain the sequence number of the lost data packet according to the SACK field;

Querying and obtaining the missing fields from the cache according to the sequence number of the data packet;

Retransmit the lost field according to the data packet sequence number.

Wherein, the field is an extension mechanism of TCP, allowing the receiver to send multiple selective acknowledgments to the sender to indicate the range of data that has been successfully received. By using the SACK field, the sender can know more precisely which data has been successfully received.

Specifically, based on the original characteristics of TCP, which data packets are lost can be confirmed. For example, TCP repeatedly sends a certain ACK or sends a SACK field, which can clearly indicate which data packets are lost.

Wherein, the historical packet loss rate can be obtained by statistics of historical data.

Further, the method also includes:

In the process of retransmitting the lost field according to the sequence number of the data packet, configuring a timer for each retransmitted lost field;

Retransmitting the lost fields for each retransmission based on the timer with a timeout.

Through the above embodiment, when retransmission is performed, a pre-configured timer can also be obtained, and retransmission with timeout is triggered based on the timer. Specifically, if the acknowledgment is not received for a long time, it can be analyzed through related functions that there is a risk of packet loss, and a timeout retransmission can be triggered at this time. For example: every time a data packet is retransmitted, a timer will be set for the data packet. If the confirmation is not received for a long time, the timer will retransmit after a timeout.

In actual scenarios, if packets with sequence numbers 1, 2, 3, and 5 are captured, packets with sequence numbers 4 may be lost, or, if packets with sequence numbers 1, 2, 3, and 4 are captured, But the data packet with the sequence number 5 has not been captured for a long time, and the data packet with the sequence number 5 may be lost.

It is understandable that network fluctuations or packet loss are often encountered in the wireless network environment. After determining that the data packet is at risk of loss or that data packet loss has occurred, it is necessary to inform the sender of the occurrence of packet loss or wait. After the timeout, the sender will retransmit, and the retransmitted data packet also needs to go through a complete link to reach the receiver, which takes a long time. Based on the characteristics of the wireless network, this embodiment shortens the process of resending packets from the last intermediate route to the receiver, that is, directly obtains the lost fields from the cache for retransmission instead of resending the data packet from the sender, greatly The time required to resend packets is reduced. Moreover, in order to ensure that the data packet reaches the opposite end successfully, a timer is maintained for each sent data packet, so as to periodically trigger timeout retransmission.

That is to say, when a data packet is lost, the traditional TCP link needs to be retransmitted from the sender, while the TCP link using the TCP acceleration service can resend the data packet from the intermediate device, shortening the retransmission path of.

When the TCP flow detects packet loss, it will reduce the sending window value and greatly reduce the TCP flow rate. The TCP acceleration service will judge the packet loss more sensitively than the original sender, so that the original TCP flow can respond to the packet loss before Packet replenishment, thereby improving the bandwidth utilization of the overall network.

This embodiment optimizes the packet loss judgment and retransmission algorithm for the wireless network, effectively improving the problem of bandwidth drop caused by bad signals in the wireless network environment.

In this embodiment, packet loss is determined dynamically, so that packets can be replenished in advance before packet loss affects the link, so as to ensure smooth link. This embodiment confirms the arrival of the data packet in advance, shields the packet loss perception of the sender, and guarantees the transmission speed of the link to the greatest extent.

S12. During the retransmission process, detect whether the data flow corresponding to the data packet is congested, and perform congestion control when it is detected that the data flow corresponding to the data packet is congested.

Packet loss generally occurs on the wireless network that is routed to the user terminal in the middle, while the wired network usually has better network quality. When packet loss occurs, the existing congestion control algorithm based on packet loss will greatly reduce the link transmission speed.

In comparison, in this embodiment, the detecting whether the data flow corresponding to the data packet is congested, and performing congestion control when detecting that the data flow corresponding to the data packet is congested includes:

The congestion index is calculated using the following formula:

;

Wherein, R represents the pre-configured theoretical bandwidth, S represents the size of the data packet, t _RTT represents the round-trip delay of the data packet, t _RTO represents the pre-configured transmission overtime limit, p represents the historical packet loss rate, b represents the number of confirmation packets of an ACK; Y represents the congestion index, bw represents the bandwidth of the TCP flow to which the data packet belongs; cwnd represents the size of the congestion window;

Get the pre-configured congestion index threshold;

When the congestion index Y is less than the congestion index threshold, it is determined that the TCP flow corresponding to the data packet is congested;

reducing the sending frequency of the analog ACK packet used for early confirmation in the TCP flow, and increasing the retransmission frequency of the data packet corresponding to the TCP flow in the cache;

Wherein, for other TCP flows with the same IP, reduce the sending frequency of the simulated ACK packets used for early confirmation in the other TCP flows, and increase the retransmission frequency of the data packets corresponding to the other TCP flows in the cache. Dynamically evaluate and regulate other TCP flows of the same IP to prevent multiple TCP flows from being blocked at the same time.

Wherein, the congestion index threshold can be customized, such as 0.86.

It can be understood that when congestion occurs, in order to avoid further congestion caused by frequent sending of simulated ACK packets, the frequency of sending simulated ACK packets is reduced, and then the bandwidth of the corresponding TCP stream is adjusted to an appropriate size; when congestion occurs, it is also It is more prone to packet loss, so the retransmission frequency is increased to reduce the packet loss rate. Dynamically adjust the frequency of early confirmation and retransmission according to the congestion situation, reduce the overall load pressure of the TCP link, and promote the entire TCP link to return to normal as soon as possible.

In the above embodiments, the dynamic congestion control algorithm is used to reduce the possibility of link congestion and improve link bandwidth utilization.

S13. After performing congestion control, detect whether the data packet needs to be acknowledged in advance.

Since the route from the sender to the intermediate route is generally a wired network connection, while the route from the intermediate route to the receiver is a wireless network, the network from the sender to the intermediate route is much better than the network from the intermediate route to the terminal (ie, the receiver). Therefore, if the data packets are confirmed in advance endlessly, a large rate difference will be formed at both ends of the network, which will cause a large number of data packets to accumulate in the cache, so the data packets need to be analyzed according to the bandwidth and packet loss rate , to determine whether to acknowledge the packet in advance.

Specifically, the detecting whether the data packet needs to be confirmed in advance includes:

Obtain the pre-configured fourth weight, fifth weight and sixth weight;

Calculate the product of the bandwidth of the TCP stream to which the data packet belongs and the fourth weight to obtain a fourth value, calculate the product of the historical packet loss rate and the fifth weight to obtain a fifth value, and calculate the congestion index multiplied by said sixth weight to obtain a sixth value;

calculating the sum of the fourth value, the fifth value and the sixth value to obtain a target value;

Get pre-configured preset thresholds;

When the target value is greater than the preset threshold, it is determined that the data packet needs to be acknowledged in advance.

In the above embodiment, the weighted sum calculation is performed in combination with the congestion index obtained in the congestion control process, the bandwidth and the historical packet loss rate to detect whether the data packet needs to be confirmed in advance, so as to reply the ACK packet in advance and resend data packets to reduce the overall delay and packet loss rate of the link.

In this embodiment, if it is determined that the data packet does not need to be confirmed in advance, the data packet is stored in the cache.

S14. When it is detected that the data packet needs to be acknowledged in advance, generate a simulated ACK packet of the data packet, and transmit the simulated ACK packet to the sender of the data packet.

A TCP stream is in the slow start phase when it is initially established. The faster the receiver replies with ACK, the faster the link speed will increase. Therefore, replying ACK in advance can shorten the time for the TCP flow to reach the maximum bandwidth.

By simulating the ACK packet, the data packet can be confirmed faster, so that the sender can send the subsequent data packet faster, thereby improving the link transmission speed.

In the above-mentioned embodiment, the link state is improved without affecting the original TCP link by adopting the way of bypass supplementary packet. Different from traditional TCP acceleration, since this embodiment does not truncate the TCP link, but supplements packets on the basis of the original link, it supports hot start and hot update.

It can be seen from the above technical solutions that the present invention is applied to the acceleration device or acceleration service deployed by the bypass, and improves the link status without affecting the original TCP link. The effect of splitting the TCP link into two segments (that is, from the sender to the deployment node of the acceleration device or acceleration service, and from the deployment node of the acceleration device or acceleration service to the receiver), limits the impact of packet loss to any one of the links In this way, the lost data packets can be retransmitted faster, and the deployment is flexible; the data packets are captured in real time, and the captured data packets are distributed to different threads to improve the concurrent processing efficiency; when the data packet loss is detected, Combined with the cache to retransmit the data packet, there is no need to resend the packet from the sender, just get the lost packet from the intermediate acceleration cache and retransmit it to the receiver, which reduces the time required for resending the packet, and advances before the packet loss affects the link Supplement the packet to ensure the smooth link; perform congestion control when the data stream corresponding to the data packet is detected to be congested, which reduces the possibility of link congestion and improves the bandwidth utilization of the link; detects whether the data packet needs to be forwarded Confirm that by replying ACK in advance and resending data packets, the equivalent two-stage rate from the sender to the deployment node of the acceleration device or acceleration service, and from the deployment node of the acceleration device or acceleration service to the receiver is balanced, reducing the link speed. Overall latency and packet loss.

As shown in FIG. 4 , it is a functional block diagram of a preferred embodiment of the bypass TCP acceleration device of the present invention. The bypass TCP acceleration device 11 includes a distribution unit 110 , a detection unit 111 , a congestion control unit 112 , and a retransmission unit 113 . The module/unit referred to in the present invention refers to a series of computer program segments that can be executed by a processor and perform fixed functions, and are stored in a memory. In this embodiment, the functions of each module/unit will be described in detail in subsequent embodiments.

In this embodiment, the bypass TCP acceleration device 11 runs on the bypass-deployed acceleration device or acceleration service, including:

The distribution unit 110 is configured to capture data packets in real time, and when capturing data packets, distribute the captured data packets to corresponding threads;

The detection unit 111 is configured to detect whether there is a risk of data packet loss, and when it is detected that there is a risk of data packet loss, retransmit the data packet in combination with the cache, and retransmit the data packet to the recipient of the data packet;

The congestion control unit 112 is configured to detect whether the data flow corresponding to the data packet is congested during the retransmission process, and perform congestion control when it is detected that the data flow corresponding to the data packet is congested;

The detection unit 111 is further configured to detect whether the data packet needs to be confirmed in advance after performing congestion control;

The retransmission unit 113 is configured to generate a simulated ACK packet of the data packet when it is detected that the data packet needs to be confirmed in advance, and transmit the simulated ACK packet to the sender of the data packet.

It can be seen from the above technical solutions that the present invention is applied to the acceleration device or acceleration service deployed by the bypass, and improves the link status without affecting the original TCP link. The effect of splitting the TCP link into two segments (that is, from the sender to the deployment node of the acceleration device or acceleration service, and from the deployment node of the acceleration device or acceleration service to the receiver), limits the impact of packet loss to any one of the links In this way, the lost data packets can be retransmitted faster, and the deployment is flexible; the data packets are captured in real time, and the captured data packets are distributed to different threads to improve the concurrent processing efficiency; when the data packet loss is detected, Combined with the cache to retransmit the data packet, there is no need to resend the packet from the sender, just get the lost packet from the intermediate acceleration cache and retransmit it to the receiver, which reduces the time required for resending the packet, and advances before the packet loss affects the link Supplement the packet to ensure the smooth link; perform congestion control when the data stream corresponding to the data packet is detected to be congested, which reduces the possibility of link congestion and improves the bandwidth utilization of the link; detects whether the data packet needs to be forwarded Confirm that by replying ACK in advance and resending data packets, the equivalent two-stage rate from the sender to the deployment node of the acceleration device or acceleration service, and from the deployment node of the acceleration device or acceleration service to the receiver is balanced, reducing the link speed. Overall latency and packet loss.

As shown in FIG. 5 , it is a schematic structural diagram of a computer device in a preferred embodiment of the bypass TCP acceleration method of the present invention.

The computer device 1 may include a memory 12, a processor 13 and a bus, and may also include a computer program stored in the memory 12 and operable on the processor 13, such as a bypass TCP acceleration program.

Those skilled in the art can understand that the schematic diagram is only an example of the computer device 1 and does not constitute a limitation to the computer device 1. The computer device 1 can be a bus structure or a star structure. The computer The device 1 may also include more or less other hardware or software than shown in the figure, or a different arrangement of components, for example, the computer device 1 may also include input and output devices, network access devices, and the like.

It should be noted that the computer device 1 is only an example, and other existing or future electronic products that can be adapted to the present invention should also be included in the protection scope of the present invention, and are included here by reference .

Wherein, the memory 12 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. . The memory 12 may be an internal storage unit of the computer device 1 in some embodiments, such as a removable hard disk of the computer device 1 . In other embodiments, the memory 12 can also be an external storage device of the computer device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, flash card (Flash Card), etc. Further, the memory 12 may also include both an internal storage unit of the computer device 1 and an external storage device. The memory 12 can not only be used to store application software and various data installed in the computer device 1, such as the code of the bypass TCP acceleration program, etc., but also can be used to temporarily store data that has been output or will be output.

In some embodiments, the processor 13 may be composed of integrated circuits, for example, may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same function or different functions, including one or more central processing units. Central Processing unit (CPU), microprocessor, digital processing chip, graphics processor and a combination of various control chips, etc. The processor 13 is the control core (Control Unit) of the computer device 1, and uses various interfaces and lines to connect various components of the entire computer device 1, and runs or executes programs or modules stored in the memory 12 (for example, executes bypass TCP acceleration program, etc.), and call the data stored in the memory 12 to execute various functions of the computer device 1 and process data.

The processor 13 executes the operating system of the computer device 1 and various installed application programs. The processor 13 executes the application program to implement the steps in the above embodiments of each bypass TCP acceleration method, for example, the steps shown in FIG. 1 .

Exemplarily, the computer program can be divided into one or more modules/units, and the one or more modules/units are stored in the memory 12 and executed by the processor 13 to complete this invention. The one or more modules/units may be a series of computer-readable instruction segments capable of accomplishing specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device 1 . For example, the computer program can be divided into a streaming unit 110 , a detection unit 111 , a congestion control unit 112 , and a retransmission unit 113 .

The above-mentioned integrated units implemented in the form of software function modules can be stored in a computer-readable storage medium. The above-mentioned software function modules are stored in a storage medium, and include several instructions to enable a computer device (which may be a personal computer, computer device, or network device, etc.) or a processor (processor) to execute the side functions described in various embodiments of the present invention. Part of the road TCP acceleration method.

If the integrated modules/units of the computer device 1 are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the present invention realizes all or part of the processes in the methods of the above embodiments, and can also be completed by instructing related hardware devices through computer programs, and the computer programs can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can realize the steps of the above-mentioned various method embodiments.

Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , random access memory, etc.

Further, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function, etc.; Use the created data etc.

The block chain referred to in the present invention is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain (Blockchain), essentially a decentralized database, is a series of data blocks associated with each other using cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify its Validity of information (anti-counterfeiting) and generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

The bus may be a peripheral component interconnect (PCI for short) bus or an extended industry standard architecture (EISA for short) bus or the like. The bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one straight line is used in FIG. 5 , but it does not mean that there is only one bus or one type of bus. The bus is configured to realize connection and communication between the memory 12 and at least one processor 13 and the like.

Although not shown, the computer device 1 may also include a power supply (such as a battery) for supplying power to various components. Preferably, the power supply may be logically connected to the at least one processor 13 through a power management device, thereby realizing Charge management, discharge management, and power management functions. The power supply may also include one or more DC or AC power supplies, recharging devices, power failure detection circuits, power converters or inverters, power status indicators and other arbitrary components. The computer device 1 may also include various sensors, bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

Further, the computer device 1 may also include a network interface. Optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which are usually used in the computer device 1 Establish a communication connection with other computer equipment.

Optionally, the computer device 1 may further include a user interface, which may be a display (Display) or an input unit (such as a keyboard (Keyboard)). Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode, Organic Light-Emitting Diode) touch device, and the like. Wherein, the display may also be appropriately referred to as a display screen or a display unit, and is used for displaying information processed in the computer device 1 and for displaying a visualized user interface.

It should be understood that the embodiments are only for illustration, and are not limited by the structure in the scope of the patent application.

FIG. 5 only shows a computer device 1 with components 12-13. Those skilled in the art can understand that the structure shown in FIG. 5 does not constitute a limitation to the computer device 1, and may include less Or more components, or combinations of certain components, or a different arrangement of components.

With reference to FIG. 1, the memory 12 in the computer device 1 stores multiple instructions to implement a bypass TCP acceleration method, and the processor 13 can execute the multiple instructions to achieve:

Capture data packets in real time, and divert the captured data packets to corresponding threads when capturing data packets;

Detecting whether there is a risk of data packet loss, and when detecting that there is a risk of data packet loss, retransmitting the data packet in combination with the buffer, and retransmitting the data packet to the receiver of the data packet;

During the retransmission process, detect whether the data flow corresponding to the data packet is congested, and perform congestion control when it is detected that the data flow corresponding to the data packet is congested;

After performing congestion control, detecting whether the data packet needs to be confirmed in advance;

When it is detected that the data packet needs to be acknowledged in advance, generate a simulated ACK packet of the data packet, and transmit the simulated ACK packet to the sender of the data packet.

Specifically, for the specific implementation method of the above instructions by the processor 13, reference may be made to the description of relevant steps in the embodiment corresponding to FIG. 1 , and details are not repeated here.

It should be noted that the data involved in this case were obtained legally.

In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

The invention is applicable to numerous general purpose and special purpose computer system environments or configurations. Examples: personal computers, server computers, handheld or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, including A distributed computing environment for any of the above systems or devices, etc. The invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including storage devices.

The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each unit may physically exist separately, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or in the form of hardware plus software function modules.

It will be apparent to those skilled in the art that the invention is not limited to the details of the above-described exemplary embodiments, but that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention.

Accordingly, the embodiments should be regarded in all points of view as exemplary and not restrictive, the scope of the invention being defined by the appended claims rather than the foregoing description, and it is therefore intended that the scope of the invention be defined by the appended claims rather than by the foregoing description. All changes within the meaning and range of equivalents of the elements are embraced in the present invention. Any reference sign in a claim should not be construed as limiting the claim concerned.

In addition, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means stated in the present invention may also be realized by software or hardware by one unit or means. The terms first, second, etc. are used to denote names and do not imply any particular order.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention without limitation. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be Modifications or equivalent replacements can be made without departing from the spirit and scope of the technical solutions of the present invention.

Claims (11)

1. A bypass TCP acceleration method, applied to acceleration devices or acceleration services of a bypass deployment, comprising:

capturing a data packet in real time, and shunting the captured data packet to a corresponding thread when the data packet is captured;

detecting whether a data packet has a loss risk or not, and when detecting that the data packet has the loss risk, executing retransmission of the data packet in combination with a cache, and retransmitting the data packet to a receiving side of the data packet;

In the retransmission process, detecting whether the data flow corresponding to the data packet is congested, and executing congestion control when the data flow corresponding to the data packet is detected to be congested;

after congestion control is executed, detecting whether the data packet needs to be confirmed in advance;

and when detecting that the data packet needs to be confirmed in advance, generating a simulated ACK packet of the data packet, and transmitting the simulated ACK packet to a sender of the data packet.

2. The bypass TCP acceleration method of claim 1, wherein said method further comprises:

acquiring intermediate equipment of the sender and the receiver, acquiring an idle port of the intermediate equipment, configuring a mirror image for the idle port, establishing connection with a mirror image server through the idle port, and deploying the acceleration equipment on the mirror image server; or alternatively

And obtaining intermediate equipment of the sender and the receiver, and deploying the acceleration service on the intermediate equipment.

3. The bypass TCP acceleration method of claim 1, wherein said offloading captured packets to corresponding threads comprises:

acquiring the packet head format of each data packet;

Determining the protocol type of each data packet according to the packet head format of each data packet;

analyzing the four-tuple of the TCP/IP of each data packet based on the protocol type of each data packet to obtain an analysis result;

dividing each data packet into a corresponding TCP stream according to the analysis result;

and configuring corresponding threads for each data packet according to the TCP stream to which each data packet belongs.

4. The bypass TCP acceleration method of claim 1, wherein said combining buffer performs retransmission of said data packet, said method further comprising:

acquiring an MTU determined when the sender and the receiver carry out TCP handshake;

dividing the unacknowledged data packet by taking the MTU as the length, and storing the divided data packet in the form of an ordered two-dimensional linked list to the cache;

when an ACK packet of any data packet in the cache is received, deleting a field of a section to which the any data packet belongs from the cache;

and when the lost data packet is inquired in the buffer memory, performing binary search within a preset length range at the tail end of the ordered two-dimensional linked list.

5. The bypass TCP acceleration method of claim 1, wherein said detecting whether there is a risk of loss of a data packet, and when detecting that there is a risk of loss of a data packet, performing retransmission of the data packet in conjunction with buffering comprises:

Acquiring the bandwidth, the historical packet loss rate and the storage time length of the TCP stream to which each data packet belongs in the cache;

acquiring a first weight corresponding to the bandwidth, a second weight corresponding to the historical packet loss rate and a third weight corresponding to the storage duration;

calculating the product of the bandwidth and the first weight to obtain a first value, calculating the product of the historical packet loss rate and the second weight to obtain a second value, and calculating the product of the storage duration and the third weight to obtain a third value;

calculating the sum of the first value, the second value and the third value to obtain the current dynamic packet loss rate of each data packet;

acquiring a preconfigured packet loss rate threshold;

when the dynamic packet loss rate corresponding to a data packet is larger than the packet loss rate threshold value, determining that the data packet has a loss risk;

acquiring a SACK field from a TCP header corresponding to the data packet;

acquiring the lost data packet sequence number according to the SACK field;

inquiring and acquiring a lost field from the cache according to the data packet sequence number;

and retransmitting the lost field according to the data packet sequence number.

6. The bypass TCP acceleration method of claim 5, wherein the method further comprises:

In the process of retransmitting the lost field according to the data packet sequence number, configuring a timer for each retransmitted lost field;

and carrying out timeout retransmission on the lost field of each retransmission based on the timer.

7. The bypass TCP acceleration method of claim 5, wherein detecting whether the data flow corresponding to the data packet is congested, and performing congestion control when detecting that the data flow corresponding to the data packet is congested, comprises:

the congestion index is calculated using the following formula:

；

wherein R represents a preconfigured theoretical bandwidth, S represents the size of the data packet, t _RTT Indicating the round trip delay of the data packet, t _RTO Representing a preconfigured transmission timeout time limit, wherein p represents the historical packet loss rate, and b represents the number of acknowledgement packets of one ACK; y represents the congestion index, bw tableShowing the bandwidth of the TCP stream to which the data packet belongs; cwnd represents the size of the congestion window;

acquiring a pre-configured congestion index threshold value;

when the congestion index Y is smaller than the congestion index threshold value, determining that the TCP flow corresponding to the data packet is congested;

reducing the sending frequency of the simulated ACK packet for early confirmation in the TCP stream, and improving the retransmission frequency of the data packet corresponding to the TCP stream in the cache;

And for other TCP streams with the same IP, reducing the sending frequency of the simulated ACK packet for the advanced acknowledgement in the other TCP streams, and improving the retransmission frequency of the data packet corresponding to the other TCP streams in the buffer.

8. The bypass TCP acceleration method of claim 7, wherein said detecting whether said data packet requires an advance acknowledgement comprises:

acquiring a fourth weight, a fifth weight and a sixth weight which are preset;

calculating the product of the bandwidth of the TCP stream to which the data packet belongs and the fourth weight to obtain a fourth value, calculating the product of the historical packet loss rate and the fifth weight to obtain a fifth value, and calculating the product of the congestion index and the sixth weight to obtain a sixth value;

calculating the sum of the fourth value, the fifth value and the sixth value to obtain a target value;

acquiring a preset threshold value which is preset;

and when the target value is larger than the preset threshold value, determining that the data packet needs to be confirmed in advance.

9. A bypass TCP acceleration apparatus, operating on a bypass deployed acceleration device or acceleration service, comprising:

the distribution unit is used for capturing the data packet in real time and distributing the captured data packet to a corresponding thread when the data packet is captured;

The detection unit is used for detecting whether a data packet has a loss risk or not, and when detecting that the data packet has the loss risk, combining the buffer memory to execute retransmission of the data packet and retransmitting the data packet to a receiving side of the data packet;

the congestion control unit is used for detecting whether the data flow corresponding to the data packet is congested in the retransmission process, and executing congestion control when the data flow corresponding to the data packet is detected to be congested;

the detecting unit is further configured to detect whether the data packet needs to be acknowledged in advance after congestion control is performed;

and the retransmission unit is used for generating a simulated ACK packet of the data packet when the data packet is detected to need to be confirmed in advance, and transmitting the simulated ACK packet to a sender of the data packet.

10. A computer device, the computer device comprising:

a memory storing at least one instruction; and

A processor executing instructions stored in the memory to implement the bypass TCP acceleration method of any one of claims 1 to 8.

11. A computer-readable storage medium, characterized by: the computer readable storage medium having stored therein at least one instruction for execution by a processor in a computer device to implement the bypass TCP acceleration method of any one of claims 1 to 8.